A compelling body of evidence indicates that signals from rod photoreceptors use multiple pathways to reach ganglion cells. In the canonical primary pathway, rod signals gain access to downstream cone circuitry through the AII amacrines which form electrical synapses (gap junctions) with ON cone bipolar and glycinergic synapses with OFF cone bipolar. In the secondary pathway, electrical synapses between rods and cones provide a direct entry for rod signals into cone circuits. However, a credible direct demonstration of rod-cone coupling has been made only in monkey. Because technical issues prevent a conventional approach to measurement of rod-cone coupling in the mouse (i.e. injections of junction-permeant tracers using microelectrodes), we developed a novel method to evaluated coupling. The method is based on transgenic, cell-specific expression of a transporter, the movement of the transported molecule through gap junctions to neighboring cells and the detection of that molecule with specific antibodies. This method eliminates the need for any physical manipulation of the live cells. Despite the widely held assumption that rod-cone coupling underlies the secondary pathway, we found no evidence of rod-cone coupling in the mouse. Thus, we propose a set of experiments to determine the generality of the canonical secondary pathway model by testing rod-cone coupling in the rabbit and to evaluate cone-cone and possible rod-rod coupling in mouse and rabbit retinas. In contrast, our data strongly support the basic tenets of the rod primary pathway. However, while the prevailing model postulates that AII amacrines express Cx36 and cone ON bipolar express Cx45, forming a 'heterotypic'electrical synapse, our data indicate more complexity. We propose there are two types of glycinergic amacrine cells, those expressing Cx45 and those expressing Cx36 and that each forms homotypic junctions with a subset of cone bipolar cells expressing the same connexin. We hypothesize that the expression of incompatible connexins is mechanism to allow segregation of amacrine- cone bipolar interactions according to cell subtype. We will determine the types of cone bipolar involved in rod primary pathway signaling and which connexins they employ. In addition, we will determine if different retinal connexins can functionally interact. PUBLIC HEALTH RELEVANCE: Our studies address fundamental questions about the neural circuitry employed by rod photoreceptors, which contribute to retinal responses over a range of light inputs from near total darkness to bright moonlight. Disorders of the neural retina are a primary cause of human blindness and a rational pursuit of therapeutic strategies requires a full understand of mammalian retinal circuitry.